@sunychoudhary
Suny Choudhary
@sunychoudhary
Building AI Security for LLMs | CEO @ LangProtect
Austin, USAJoined March 2026
About
I’m the CEO & Co-Founder of LangProtect, where I build security and governance for applications powered by LLMs.
I write about AI security, prompt injection, data leakage, and real-world risks in production LLM systems; along with practical ways to secure them.
Currently focused on helping developers and enterprises ship AI features safely without compromising control, privacy, or trust.
Available for
Writing, speaking, and collaborating on AI security, LLM safety, and developer tooling.
Suny Choudhary's blogs
Recently published
Running Multiple LLMs in Production? Let’s Talk About the Security Gaps Nobody Mentions
3d ago · 6 min read · On paper, your system looks solid. You have multiple LLMs running in production. A chatbot for support, a copilot for internal teams, maybe a RAG pipeline pulling in company data. The APIs are secured
Join discussion
AI Data Leaks: Why Real-Time Prompt Filtering Is the Key to Prevention
Apr 8 · 8 min read · When people hear “data leak,” they usually think of a breach. An attacker gets into a system. A database is exposed. A file is stolen. Logs light up. Security responds. That is not how most AI data le
Join discussion
Prompt Injection Explained: The Biggest AI Security Risk Developers Ignore
Apr 1 · 9 min read · AI security conversations often start in the wrong place. Most teams focus on model choice, response quality, latency, or cost. Those things matter. But they are not the first place real security fail
Join discussion